This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elseviers archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights

Author's personal copy

Biochemical Engineering Journal 79 (2013) 8493

Contents lists available at ScienceDirect

Biochemical Engineering Journal

journal homepage: www.elsevier.com/locate/bej

Regular article

Development of simplied anaerobic digestion models (SADMs) for

studying anaerobic biodegradability and kinetics of complex biomass
O.L. Yusuf Momoh a, , B.U. Anyata b,1 , D.P. Saroj c,2
a

Department of Civil and Environmental Engineering, University of Port Harcourt Choba, P.M.B 5323 Rivers State, Nigeria
Department of Civil Engineering, University of Benin, P.M.B 1154 Benin City, Edo State, Nigeria
c
Department of Civil and Environmental Engineering, University of Surrey, Surrey GU2 7XH, United Kingdom
b

a r t i c l e

i n f o

Article history:
Received 20 March 2013
Received in revised form 26 June 2013
Accepted 30 June 2013
Available online 12 July 2013
Keywords:
Anaerobic process
Biodegradability
Biogas
Kinetic parameters
Growth kinetics
Rate limiting

a b s t r a c t
The anaerobic co-digestion of cow manure and waste paper at ambient temperature condition was
observed to be optimized at a mix proportion of 75:25 respectively. The development and testing of
a set of simplied anaerobic digestion models (SADMs) for this mixture revealed that the Hills based
biogas yield rate model was most appropriate in describing the kinetics of biogas production. Parameter estimation using non-linear regression revealed that the half saturation constants expressed as
acidied substrate and volatile solids equivalents were 0.228 g/L and 5.340 g VS/L respectively, and the
maximum specic biogas yield rate and biodegradability were 2.2 mL/g VS/day and 0.313 respectively.
The coefcients n and m indicative of acidogenic bacterial adaptation for degradation and acetogenic/methanogenic bacterial cooperativity were estimated to be 1.360 and 2.738 respectively, while
hydrolysis/acidogenesis was considered the rate limiting step. The need of bacterial adaptation may be
an important factor to consider during anaerobic modeling of complex biomass.
2013 Elsevier B.V. All rights reserved.

1. Introduction
Anaerobic processes in waste management have been widely
applied on account of their operational simplicity and potential
of energy recovery [1]. Anaerobic digestion is the breakdown of
organic material to produce biogas which is a mixture of methane
and carbondioxide that is catalyzed by a consortium of microorganisms in a series of interlinked biochemical reactions. These
biochemical reactions comprise of hydrolysis, acidogenesis, acetogenesis and methanogenesis. Hydrolysis is the breakdown of
complex biomass into monomeric units; acidogenesis is the conversion of the monomers into volatile fatty acids; acetogenesis
is the conversion of the volatile fatty acids into acetic acid and
methanogenesis is the conversion of acetic acid into methane and
carbondioxide [2,3]. Anaerobic digestion has been identied as not
only a viable means of producing carbon neutral energy [4] but also
a means of mitigating the adverse effect of uncontrolled greenhouse
gas emissions during decay of organic matter in the environment
[5].

Corresponding author. Tel.: +234 80 3538 6779.

E-mail addresses: yusuf.momoh@uniport.edu.ng (O.L.Y. Momoh),
bufanyat a@yahoo.co.uk (B.U. Anyata), d.saroj@surrey.ac.uk (D.P. Saroj).
1
Tel.: +234 7067538129.
2
Tel.: +44 01483686634.
1369-703X/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.bej.2013.06.018

Organic substrate utilized for anaerobic digestion range from

wastewater to complex organic feed stock such as animal manure,
agricultural and industrial waste [6]. However, more recently, the
process of co-digesting complex feedstock has been reported to
result in improved biogas yield [7,8]. Thus, for proper utilization
of raw material in anaerobic digestion, adequate understanding
of anaerobic biodegradation kinetics is imperative. Although, the
anaerobic biodegradation kinetics of wastewater is well established, it has been poorly developed for complex biomass due to
various reasons.
Most of the models used for studying biodegradation kinetics
are based on maximum specic growth rate (max ) which requires
short retention time that is not feasible for the complex biomass. In
addition, the differentiation between bacteria volatile suspended
solids and complex biomass volatile solids can be very difcult
[9,10]. Also, most of the models currently in use are based on a
soluble growth limiting substrate whereas complex biomass exists
in non-soluble form [11], in addition, the presence of recalcitrant
fractions in complex biomass can render some of the volatile solids
unavailable for bacteria, thus providing a false measure of the available substrate [11].
Most of the earlier available models used for anaerobic digestion
did not account for the complex nature of the natural feedstock
material because the substrate was assumed to be homogenous
and biodegradable. However, in situations where models attempt
to account for the nature of complex feedstock, they were restricted

Author's personal copy

O.L.Y. Momoh et al. / Biochemical Engineering Journal 79 (2013) 8493

Nomenclature
Af

rate limiting step coefcient for fast substrate utilization

As
rate limiting step coefcient for very slow substrate
utilization
rate limiting step coefcient for fast or very slow
Af(s)
substrate utilization
b
fraction of initial volatile solids remaining in efuent
Monods half saturation constant for acidied subks
strate (g/L)
Ks
Monods half saturation constants in volatile solids
equivalents (g/L)
kn
Hills half saturation constant for acidied substrate
(g/L)
Kn
Hills half saturation constant in volatile solids
equivalents (g/L)
substrate inhibition constant for acidied substrate
ki
(g/L)
coefcient of acetogenic/methanogenic bacteria
m
adaptation for cooperativity
n
coefcient of acidogenic bacteria adaptation for
complex substrate degradation
recalcitrant fraction
Rf
Rmax
maximum specic biogas yield rate (mL/g VS/day)
R
specic biogas yield rate (mL/g VS/day)
S0
initial volatile solids concentration (g/L)
volatile solids concentration remaining (g/L)
S
X(a)
acidogenic biomass concentration (mass/volume)
X(a/m)
acetogenic/methanogenic biomass concentration
(mass/volume)
concentration of acidied substrate generated (g/L)
Sh
Sh(i)
concentration of acidied substrate remaining
intracellularly
Sh(u)
concentration of utilized acidied substrate by the
acetogenic/methanogenic biomass (g/L)

bacteria growth rate (/day)
max(a) maximum acidogenic bacteria growth rate (/day)
max(a/m) maximum acetogenic/methanogenic bacteria
growth rate (/day)
Yx/s(a)
yield coefcient for acidogenic biomass production
(g/L)/(g VS/L)
Yx/s(a/m) yield coefcient for acetogenic/methanogenic
biomass production (g/L)/(g VS/L)
Yy/s
yield coefcient for biogas production (mL/g VS)/
(g VS/L))
KH(a)
maximum substrate utilization rate by acidogenic
bacteria (g VSutilized /L/day)
KH(a/m) maximum substrate utilization rate by acetogenic/
methanogenic bacteria (g acidied substrateutilized /
L/day)
yt
biogas yield (mL/g VS)
time (day)
t

85

are the most appropriate tool for studying the operation and technology development of anaerobic process [1,17].
For designing of anaerobic processes, simplied model are considered most appropriate [1,18]. Although, the need for a two
stage model comprising hydrolysis and uptake of hydrolyzed
substrate has been viewed by Batstone [1] as more appropriate
for designing anaerobic processes, simplied generalized models
based on rst order models involving single stage have predominantly been employed in designing anaerobic system involving
complex biomass. Recently, Linke [19] and Momoh and Nwaogazie
[3] applied a rst order biogas yield model in sizing continuous
stirred tank and batch reactors respectively. In addition, the development of simplied kinetic models for more specic waste such
as, organic fraction of municipal solid waste (OFMSW) has been
reported, however, these simplied models were based on maximum specic bacteria growth rate (max ) [20,21].
In this study, a set of simplied kinetic models has been formulated by applying the biogas yield approach. This approach allows
the estimation of various parameters such as recalcitrant fraction,
biodegradable fraction, biodegradability and maximum biogas production rate. The model predictions have been assessed against the
experimental biogas yield obtained by using representative samples of complex biomass.
2. Kinetic model development
The process of studying bacterial growth kinetics has been
largely followed using the classical Monod growth kinetic model
[22]. Though, this model has been established to be more appropriate in describing the growth process for pure culture utilizing
homogenous substrates than for heterogeneous culture utilizing
heterogeneous substrate [11,23], signicant amount of studies on
the kinetics of microbial growth and biodegradation involving
mixed culture and complex substrates are still been described using
the Monod growth kinetic model [22].
The heterogeneous nature of bacterial and the complex nature
of substrate utilized in this study necessitated the consideration
of other bacteria growth models such as the Mosers growth model
[22] and its homologue, the Hills growth kinetic model as proposed
by Liu [22] and other inhibition models.
The model development involved the aggregation of hydrolysis/acidogenesis and acetogenesis/methanogenesis processes, and
the process of biogas production was assumed to comprise (i)
hydrolysis/acidogenesis by acidogenic bacteria to produce acidied
substrate for the acetogenic/methanogenic; (ii) uptake of acidied
substrate by acetogenic/methanogenic bacteria and (iii) acidied
substrate assimilation, growth and biogas production by the acetogenic/methanogenic bacteria. In this modeling approach, the
substrate utilization model of Graus [23] was used to describe the
kinetics of the rst two steps, while the process of substrate assimilation and growth of the acetogenic/methanogenic bacteria was
studied by testing the Monod, Moser, Hill and Haldanes growth
models.
2.1. Modeling hydrolysis/acidogenesis and growth of the
acidogenic bacteria

to particular substrate such as liquid manure [12,13], sewage sludge

[14] or biological waste [15].
Recently, the utilization of advance models which require extensive characterization of feedstock utilized in anaerobic digestion
has dominated literature. The anaerobic digestion model no. 1
(ADM 1) [2] and the model developed by Angelidaki et al. [16]
are examples of generalized models used for studying anaerobic
digestion of complex and co-digested biomass respectively. These
models are rigorous and require large input parameters. Also, they

The process of modeling the hydrolysis step using the rst

order kinetic model as reported by Eastman and Ferguson [24]
has been described to be unsuitable for studying the digestion
of complex biomass of co-digested substrate or more complex
biomass [2527]. In order to provide an appropriate description
of the kinetics of hydrolysis, several researchers have modied the
rst order kinetic model as developed by Eastman and Ferguson
[24].

Author's personal copy

86

O.L.Y. Momoh et al. / Biochemical Engineering Journal 79 (2013) 8493

Sanders et al. [28] developed a surface based kinetic model to

describe the disintegration of complex substrate, however, this
model failed to account for the recalcitrant fractions because it
assumed the entire substrate to be biodegradable. Modication of
the Sanders disintegration model was conducted by Esposito et al.
[26] to describe the disintegration of organic fraction of municipal
solid waste co-digested with sewage sludge. However, extensive
characterization of complex biomass in terms of various characteristics, such as particle size distribution, carbonhydrates, proteins,
lipids and inert was required for modeling the anaerobic process.
In this study, a simple substrate characterization model development was conducted that could provide an estimate of the
recalcitrant fraction of complex biomass. Hydrolysis and acidogenesis were lumped together and they were assumed to be catalyzed
by acidogenic bacteria releasing extracellular enzymes that are
adsorbed on the surface of the complex biomass. The model of
Grau [22,23,29] represented by Eq. (1a) which was subsequently
modied as represented by Eq. (2) was adopted for modeling
hydrolysis/acidogenesis.
dS

=
dt

max(a) X(a)



Yx/s(a)

S
S0

n
(1a)

where (dS/dt) represents the rate of change of complex substrate,

and S represent the concentration of complex biomass volatile
solids concentration remaining in efuent. S0 represents the initial
complex biomass volatile solid concentration, max(a) represents
the maximum growth rate of acidogenic bacteria, Xa represents the
acidogenic bacteria concentration, while Yx/s(a) represents the acidogenic bacteria biomass yield coefcient. However, in this study
(X(a) ) was assumed to be constant such that Eq. (1a) could be rewritten as

 S n

dS
= KH(a)
S0
dt

(1b)

where KH(a) is the maximum substrate utilization rate

(g VSutilized /L/day) for the acidogenic bacteria and n is the
coefcient or degree of acidogenic bacteria adaptation for complex
substrate degradation. Vavilin et al. [25] emphasized the importance of considering the recalcitrant fraction of complex biomass
(Rf ) when modeling hydrolysis of complex biomass thus, upon
considering the recalcitrant fraction of complex biomass, Eq. (2)
can be expressed as follows

S S0 Rf
dS
= KH(a)
S0
dt

n

(2)

The term KH(a) is a measure of the maximum rate of volatile

solids utilization by the acidogenic bacteria to produce acidied
substrate while, the term n, could be viewed as a measure of
the degree of volatile solids degradation by the acidogenic bacteria
which, is largely dependent on the degree of bacteria adaptation. When n equals unity, the bacteria enzyme concentration
is assumed to be in excess and the need for bacteria to adapt is
not a pre-requisite for degradation. However, when n is greater
than unity, enzyme concentration is assumed to be low such that,
adaptation of the hydrolytic/acidogenic bacteria is a necessary prerequisite for complex biomass degradation. The value of n less
than unity implies a poorly adapted acidogenic bacteria population
and as the value of n approaches zero the reaction rate becomes
independent of substrate concentration.
However, Eq. (2) can be expressed in terms of acidied substrate
produced from the complex biomass as follows
dSh
= KH(a)
dt



S S0 Rf
S0

n

Sh
S0

is a modied form of the hydrolytic step as utilized by Barthakur

et al. [11]; Faisal and Unno [30]; and Zinatizadeh et al. [31].
2.2. Uptake of acidied substrate by acetogenic/methanogenic
bacteria
The uptake or utilization of the acidied substrate into acetogenic/methanogenic biomass was modeled using the rst order
Graus substrate utilization model. In this modeling approach,
uptake or utilization rate was considered to be inversely
proportional to the initial volatile solids concentration; and
directly proportional to the concentration of the active acetogenic/methanogenic biomass which was assumed to be constant
(X(a/m) , g/L) [23]; and also directly proportional to the difference
in concentration of the acidied substrate generated by the acidogenic bacteria (Sh ) and that remaining inside the cell of the
acetogenic/methanogenic bacteria biomass (Sh(i) ), such that, the
substrate utilization rate of the acidied substrate by the acetogenic/methanogenic bacteria biomass can be represented by Eqs.
(4b)(4e).
dSh(u)
dSh(i)
dSh
=
+
dt
dt
dt
Re-arranging Eq. (4a)
dSh(u)
dt

where Sh represents the concentration of acidied substrate solubilized from the complex biomass. It is worthy to note, that Eq. (3)

dSh(i)
dSh

dt
dt

(4b)

Expressing Eq. (4b) in terms of a rst order Graus model one

obtains

dSh(u)
dt

max(a/m) X(a/m)

(Sh Sh(i) )

Yx/s(a/m) S0

(4c)

This can be re-written as Eq. (4d) at constant acetogenic/methanogenic bacteria concentration (Xa/m )

dSh(u)
dt

KH(a/m)

S0

Sh Sh(i)

(4d)

where KH(a/m) (g/L/day) is the maximum substrate utilization rate

by the acetogenic/methanogenic bacteria while (Sh Sh(i) ) represents the concentration of acidied substrate taken up by the
acetogenic/methanogenic bacterial biomass.
Expressing Sh(i) in terms of Sh , Eq. (4d) can be re-written as

dSh(u)
dt

KH(a/m)
S0

(Sh Sh )

(4e)

where is the fraction of Sh remaining intracellularly inside the

acetogenic/methanogenic biomass if the hydrolyzed acidied substrate is not metabolized very fast, due to presence of inhibitory
substances leading to accumulation of organic acids [32].
However, because the hydrolyzable/acidied substrate produced in Eq. (3) serves as substrate for the acetogenic/
methanogenic bacteria biomass that was utilized fast enough, the
intracellular concentration of the acidied substrate was assumed
to be negligible (Sh(i) = 0),
Such that Eq. (4a) can be re-written as


KH(a)

(3)

(4a)

S S0 Rf
S0

n

Hence,

Sh =

S0 KH(a)
KH(a/m) + KH(a)

Sh
S0



KH(a/m)
S0

S Rf S0
S0

Sh

(5)

n
(6a)

Author's personal copy

O.L.Y. Momoh et al. / Biochemical Engineering Journal 79 (2013) 8493

Expressing S as a function of the initial inuent volatile solids

concentration Eq. (6a) can be re-written as

Sh =

S0 KH(a)

KH(a/m) + KH(a)

(b Rf )

(6b)

where, b is the fraction of the initial substrate volatile solids concentration remaining in the efuent (that is, S = bS0 )
Assuming
KH(a)
KH(a/m) + KH(a)

= Af

(7)

where Af , represents the rate limiting step coefcient or solubilization fractional efciency for the anaerobic process in
which, the acidied substrate are metabolized very fast by the
acetogenic/methanogenic bacteria such that, uptake by the acetogenic/methanogenic bacteria is not considered the rate limiting
step. The rate limiting coefcient or solubilization fractional efciency (Af ) can be viewed as a ratio between the maximum
substrate utilization rate for production of acidied substrate by
the acidogenic bacteria to the sum of the maximum substrate utilization rate for both the acidogenic and acetogenic/methanogenic
bacteria population, such that, (Af ) may be expected to range from
0 to 1.
Thus, Eq. (6b) may be re-written as
Sh = S0 Af (b Rf )

(8)

The term (S Rf S0 ) represents the biodegradable substrate

present in the volatile solids, however not all of these fractions
are hydrolysable for uptake by the acetogenic/methanogenic bacteria cells due to environmental factors. Hence, (Sh ) represent the
actual amount of the substrate that was acidied and utilized by
the acetogenic/methanogenic bacteria.
However, conditions may exist where the substrate utilization rate by the acetogenic/methanogenic bacteria become very
slow such that the maximum substrate utilization rate by the
acidogenic bacteria for production of acidied substrate becomes
higher than the maximum substrate utilization rate for utilization of the acidied substrate by the acetogenic/methanogenic
bacteria. For example, the acidic nature of the acidied substrate
produce in excess can lead to decrease in pH, because high production of acidied intermediates can dissociate to produce protons
which can compromise the neutral pH conditions required by the
methanogenic bacteria for optimum performance [33], such that,
Sh(i) =
/ 0.
Thus, Eq. (8) can be expressed as Eq. (9) for very slow utilization
of hydrolyzed acidied substrate by the acetogenic/methanogenic
bacteria.

Sh =

S0 KH(a)
KH(a/m) (1 ) + KH(a)

(b Rf )

KH(a)
KH(a/m) (1 ) + KH(a)

2.3. Assimilation and growth of acetogenic/methanogenic

bacteria
In modeling the assimilation and growth process of the acetogenic/methanogenic step, the growth models of Monod, Moser,
Hill, and Haldane were considered. Although, the growth model
of Monod has predominantly been used to describe growth processes for low substrate concentration, the possibility of a Mosers
and more recently, the Hills growth model as proposed by Liu [22]
to describe growth kinetic at low substrate concentration had to
be considered because of the complex nature of the substrate and
mixed culture of micro-organism. For growth processes affected
by acidity of the acidied substrate, the growth model of Haldane
(Andrews) was employed [34]. Also for growth process affected by
the allosteric effectors present in the acidied substrate, the Haldane (Non-competitive) model as described by Noykova et al. [35]
was utilized to describe the growth process.
The acetogenic/methanogenic bacteria have been reported to
have a minimum doubling time of about 14 days [36], however,
for the sake of simplicity, these bacteria were lumped together.
The process following assimilation of acidied substrate led to
cell growth of acetogenic/methanogenic bacteria and production
of biogas (methane and carbondioxide). Similar process of lumping
acetogenic/methanogenic bacteria was reported by Vavilin et al.
[37] and Vavilin and Angelidaki [38] for modeling anaerobic digestion of solid waste.
The yield coefcient for biogas yield has been represented as
Yy/s =

(10)

where As is the rate limiting coefcient for very slow substrate

utilization by the acetogenic/methanogenic bacteria elicited by
presence of inhibitors which could be the acidied substrate in
excess or other substances present in the acidied substrate.
It is worthy of note that, when the maximum substrate
utilization rate for the acidogenic bacteria (KH(a) ) is less than
that of the maximum substrate utilization rate for the acetogenic/methanogenic bacteria (KH(a/m) ), the rate limiting coefcient
becomes less than 0.5 thus, implying hydrolysis/acidication as

dyt
dS

(11)

And, the yield coefcient for biomass production was represented as

Yx/s(a/m) =

dX(a/m)

(12)

dS

The Monod growth model for the acetogenic/methanogenic

bacteria can be represented by
=

= As

the rate limiting step. However, if the maximum substrate utilization rate for the acidogenic bacteria (KH(a) ) is greater than
that of the maximum substrate utilization rate for the acetogenic/methanogenic bacteria (KH(a/m) ), the rate limiting coefcient
becomes greater than 0.5, such that, acetogenesis/methanogenesis
is considered the rate limiting step. In addition, if the maximum
substrate utilization rate for the acidogenic bacteria (KH(a) ) is equal
to that of the maximum substrate utilization rate for the acetogenic/methanogenic bacteria (KH(a/m) ), the rate limiting coefcient
becomes equal to 0.5.

(9)

Hence, the rate limiting coefcient for very slow substrate

utilization of acidied substrate by the acetogenic/methanogenic
bacteria can be written as

87

max(a/m) Sh

(13)

kS + Sh

But acetogenic/methanogenic bacteria growth rate can be represented as

dX
= X(a/m)
dt

(14)

However, the yield coefcient for acetogenic/methanogenic

bacteria growth can be expressed as
dX/dt
= Yx/s(a/m)
dS/dt

(15)

Thus, substrate utilization rate can be represented as

1
dS
=
Yx/s(a/m)
dt

 dX 
dt

(16)

Author's personal copy

88

O.L.Y. Momoh et al. / Biochemical Engineering Journal 79 (2013) 8493

Hence, Eq. (16) can be re-written as

X(a/m) max(a/m) Sh
1
dS
=
Yx/s(a/m)
dt
kS + Sh

(17)

Similarly, the yield coefcient for biogas yield by the acetogenic/methanogenic bacteria can be represented as
dyt /dt
= Yy/s
dS/dt

(18)

Such that, the biogas yield rate can be represented as

dS
dyt
= Yy/s
dt
dt

(19)

Substituting Eq. (17) into Eq. (19) one obtains

Yy/s X(a/m) max(a/m) Sh
dyt
=
Yx/s(a/m)
dt
kS + Sh

Yy/s KH(a/m) Sh

(21)

ks + Sh

R=

Rmax S0m
(knm /Am
(b Rf )
f

mn

(27)

) + S0m

It is important to note that kn represents the Hills half saturation constant and kn /Af (b Rf )n represents the Hills half saturation
constant in volatile solids equivalent which can be represented as
Kn.
In cases where the acidic nature affects the utilization of acidied substrate, the Haldanes (Andrews) growth model [34,39]
represented by Eq. (28) was employed to describe bacteria growth.

(20)

It is important to note that the growth rate of the acetogenic/methanogenic bacteria was assumed to be very slow or relatively constant such that max(a/m) X(a/m) /Yx/s(a/m) was replaced with
the term KH(a/m) (g VSutilized /L/day) which represent the maximum
substrate utilization rate by the acetogenic/methanogenic bacteria.
Additionally, the death rate of the acetogenic/methanogenic bacteria (kd , /day) was assumed to be negligible due to the slow growth
rate of these micro-organisms. Furthermore, the multiplication of
Yy/s ((mLbiogas /g VS)/(g VSutilized /L)) and KH(a/m) resulted in the maximum specic biogas yield rate (Rmax ) (mLbiogas /g VS/day), while
dyt /dt (mLbiogas /g VS/day) can be described as the specic biogas
yield rate (R) at the end of biogas production.
Thus, Eq. (20) can be re-written as
R=

The Hills based biogas yield rate model was developed as represented by Eq. (27) by following similar derivation as conducted
for the Monod based biogas yield rate model.

=

max(a/m) Sh

(28)

Sh + kS + (Sh2 /ki )

In this growth process, the acidic nature of acidied substrate

may affect its metabolism such that substrate utilization is slow but
not necessarily the rate limiting step (Af ) or very slow to become the
rate limiting step (As ). Thus, the Haldanes (Andrews) based biogas
yield rate model can be represented by Eq. (29).
R=

Rmax S0
n

S0 + (kS /Af(s) (b Rf ) ) + ((S02 )(Af(s) (b Rf ) )/ki )

(29)

The Haldane (non-competitive) growth rate model assumes that

the acidied substrate may non-competitively affect growth process through allosteric mechanisms. Haldane (non-competitive)
growth rate model can be described by Eq. (30) [35,40].

Hence, by substituting Eq. (8) into Eq. (21) one obtains

R=

Rmax Af S0 (b R)
ks + Af S0 (b R)

(22)

Eq. (22) can be use to describe the biogas yield rate from complex
biomass considering acidied substrate as limiting. However, Eq.
(22) can be re-arranged so that the volatile solids apparently appear
to be the limiting substrate as represented by Eq. (23)
R=

Rmax S0
n

(kS /Af (b Rf ) ) + S0

(23)

The term ks represent the Monod half saturation constant for

the acidied substrate while ks /Af (b Rf )n represents the Monod
half saturation constant in volatile solids equivalent which can be
represented as Ks .
Similar process was applied to develop the Mosers based biogas
yield rate model by assuming that the growth process of the acetogenic/methanogenic bacteria can be described using the Mosers
growth model represented by Eq. (24).
=

max a/m Snm

(24)

kS + Snm

=

max(a/m) Sh

In this form, the allosteric nature of the acidied substrate

may affect its metabolism such that substrate utilization is slow
but not necessarily the rate limiting (Af ) or very slow to become
the rate limiting step (As ). Here, the afnity for the acidied substrate is not affected but its utilization is hindered [41]. Thus, the
non-competitive Haldane based biogas yield rate model can be represented as
R=

Rmax S0 Af(s) (b Rf )

R=

(b Rf )
(kS /Am
f

mn

) + S0m

(25)

where, m represents the degree of acetogenic/methanogenic bacterial adaptation for cooperativity, which should always be greater
than unity (m > 1) as described by Moser [22]. Again, the Mosers
growth kinetic model can be re-arranged to appear as a Hills function as proposed by Liu [22] represented by Eq. (26),
=

max(a/m) Shm
knm + Shm

(26)

(kS + S0 Af(s) (b Rf ) )(1 + (S0 Af(s) (b Rf ) /ki ))

(31)

The various model parameters were evaluated using the solver

function of the Microsoft Excel tool Pak and the most appropriate
models were selected based as their high correlation coefcient
and low root mean square error (RMSE). In situations where more
than one model share similar correlation coefcient and RMSE, the
second-order Akaikes information criterion (AICc ) was employed
to compared these models [39,42].

Thus, the Mosers based biogas yield rate model becomes

Rmax S0m

(30)

(Sh + kS )(1 + (Sh /ki ))

AICc = 2K + n log

 SS

reg
n

2K(K + 1)
n K 1

(32)

where
SSreg is the residual sum of square represented by

[di f(x)]2 and di is the experimental data while f(x) is the estimated data of the tted model [39]. The number of available points
was represented by n*, while, K represented the number of parameter to be estimated. When the difference in AICc between two
models is less the 2, no difference is believed to exist between the
models thus, both models could be used to represent the given data
points [39].

Author's personal copy

O.L.Y. Momoh et al. / Biochemical Engineering Journal 79 (2013) 8493

The raw material utilized in this study comprised cow manure

and waste paper. Cow manure was obtained from abattoir situated at Choba Community, Rivers State (Nigeria) and waste paper
was obtained from dumpsites situated at the University of Port
Harcourt, Rivers State Nigeria. About 500 g of cow manure was collected and sun dried at ambient temperature for a period of 20 days;
it was subsequently crushed using a mortar and pestle and about
500 g of waste paper was sun dried which was afterwards ground to
ne particles using a grinding mill. The volatile solids content and
carbon to nitrogen ratio were determined according to APHA [43].
Volatile solids for cow manure and waste paper were determined
to be 66.1% and 85.7% respectively using a mufe furnace, Carbolite model LMF 4 manufactured in England, and carbon to nitrogen
ratio was determined to be 22:1 and 150:1 respectively.

biogas yield

200
180
160
140
120
100
80
60
40
20
0

specic biogas yield rate

2.5
2
1.5
1
0.5

specic biogas yield rate

(mL/g VS/day)

3.1. Substrate collection

biogas yield (mL/g VS)

3. Materials and methods

89

0
0

4
6
Total solids concentarion (%)

10

Fig. 1. Biogas yield and specic biogas yield against total solids concentration.

also conducted in duplicates and allowed to run at average ambient

temperature of 28 4 C.

3.2. Experimental methodology

4. Results and discussion

In this approach of studying the anaerobic biodegradability of

complex biomass, the experimental work was conducted in two
phases. The rst phase was designed to optimize the substrate mix
proportion of cow manure and waste paper and the second phase
was designed to maximize biogas production from the optimal mix
proportion obtained in the rst phase of the experimental work.

In the rst phase, a retention time of about 40days was maintained in almost all the digesters studied, and the mixture of cow
manure and waste paper combined in the proportion of 75:25
(A2) produced the highest quantity of biogas (921 12 mL) and the
methane content was determined to be 58 3% or 534.15 12 mL
of methane (Table 1). The batch digester comprising of cow manure
alone (A1) produced 421 10 mL of biogas with methane content
of 52 2% or 218.92 10 mL of methane. The digesters A3, A4, and
A5 had insignicant quantity of methane in the biogas produced.
The low methane content in these digesters could be attributed to
shock or instability due to high volatile acid formation following
the hydrolysis of waste paper. The high performance of digester
A2 strongly underscores the benets of co-digestion in this study
which may include reduced toxicity, nutrient balance and microbial
synergism [8].
In the second phase of the experiment, the process of maximizing biogas yield from this optimal mix of cow manure and waste
paper (75:25) determined in this study was conducted in nine (9)
digesters that comprised total solids ranging from 1 to 9%. After
80 days retention time, the biogas yield and specic biogas yield
rate were observed to increase as substrate concentration increased
from 1 to 4%, but remained almost steady for substrate concentration from 5 to 9% (Fig. 1). However, digester B3 exhibited difculty
in producing signicant amount of biogas and hence it was eliminated from the study. The longer retention time experienced in
the second phase may be attributed to a reduced average ambient
temperature of 28 4 C.

3.2.1. Experimental procedure for substrate optimization

The experiment was conducted using ve Buchner asks operated in a batch mode. A split plot design approach as utilized by
Shin et al. [44] comprising a total of 5 treatments of cow manure
and waste paper were mixed in the ratio of 100:0 (A1), 75:25 (A2),
50:50 (A3), 25:75 (A4) and 0:100 (A5). The substrates were loaded
in the Buchner asks each with volumetric capacity of 500 mL containing 250 mL of water and corked to exclude air. The experiments
were conducted in duplicates and were allowed to run at an average ambient temperature of 30 3 C and the pH of the digesters
are as shown in Table 1. The biogas produced was measured by
water (brine solution) displacement method and agitation of the
batch reactors was carried out twice daily. The biogas produced was
analyzed using Gas Chromatography Agilent Technologies Model
1890A. The total solids content loaded in all digesters was xed at
6.5% which was within the recommended range of 412% for low
solid loading anaerobic digestion [45].

3.2.2. Experimental procedure for biogas maximization

In order to maximize biogas production from the optimized substrate mix obtained in the experiment described above, nine sets of
batch digesters comprising cow manure and waste paper mixture
in proportion of 75:25 were set up in batch digesters labeled B1,
B2, B3, B4, B5, B6, B7, B8 and B9, which consisted of total solids
concentration 1, 2, 3, 4, 5, 6, 7, 8 and 9% respectively. The digesters
were setup as described by Momoh and Nwaogazie [3] and were

4.1. Kinetics and biodegradability parameter estimation and

model validation
The process of characterizing the optimal mix proportion of cow
manure and waste paper (75:25) involved the application the biogas yield rate models of Monod, Moser, Hill and Haldanes models

Table 1
Digester characteristics and biogas composition.
Digester

Mix proportion

Weight of cow
manure (g)

Weight of
waste paper (g)

Conc. volatile
solids (g/L)

pH

A1
A2
A3
A4
A5

100:0
75:25
50:50
25:75
0:100

17.40
13.05
8.70
4.35
0.00

0.00
4.35
8.70
13.05
17.4

46.00
49.36
52.80
56.20
59.64

7.3
7.3
7.2
7.2
7.1

Cumulative
biogas (mL)

0.04
0.02
0.03
0.04
0.03

421
921
164
152
260

10
12
22
23
34

CH4 (%)
52
58
10
9.0
12

2
3
3
3
3

CO2 (%)
48
42
90
91
88

2
3
3
3
3

Author's personal copy

90

O.L.Y. Momoh et al. / Biochemical Engineering Journal 79 (2013) 8493

as illustrated in this study. The kinetic and biodegradability parameters estimated in this study include;
(a) Monod half saturation constant for the acidied substrate (ks )
(g/L).
(b) Monod half saturation constant in volatile solids equivalent
(Ks ) (g/L).
(c) Hills half saturation constant for the acidied substrate (kn )
(g/L).
(d) Hills half saturation constant in volatile solids equivalents (Kn )
(g/L).
(e) Maximum specic biogas yield rate (Rmax ) (mL/g VS/day)
(f) The coefcient m
(g) The coefcient n
(h) Fraction of volatile solid remaining in efuent (b)
(i) The recalcitrant fraction (Rf ).
(j) Fraction of biodegradable volatile solids (1 Rf )
(k) Fraction of biodegradable volatile solids remaining in efuent
(b Rf )
(l) Biodegradability (1 b)
(m) Rate limiting coefcient for fast or very slow uptake of acidied
substrate (Af(s) )
The results of parameter estimation using non-linear regression
are presented in Table 2. It was observed that the ve models tested
in this study can be utilized to characterize anaerobic biodegradation kinetics because each provided a high correlation coefcient
(r) of 0.99. However, the process of selecting the most appropriate model resided in the observance of the root mean square error
(RMSE). Models with the lowest root mean square error (RMSE) are
normally considered more appropriate to describe a given data set
if they share similar correlation coefcient.
The ve models tested in this study produced correlation coefcient (r) of 0.99 each, however, the Moser and Hills based biogas
yield rate models provided the lowest root mean square error
(RMSE) of 5.87E03 each, while the Monod, Haldane (Andrews),
Haldane (non-competitive) based biogas yield rate models provided higher RMSE of 0.0428, 0.0256 and 0.0256 respectively. Thus,
only the Moser and Hills based biogas yield rate models were considered most appropriate in describing the specic biogas yield rate
from this biomass mixture because they provided the least RMSE.
However, because these selected models produced similar
correlation coefcient (r) and RMSE, a second-order Akaikes
information criterion (AICc ) [42] was employed to assess model
superiority. Upon computation, the second-order Akaikes information criterion analysis produced again, similar AICc value of 97.33
each, for both models (Table 3) implying that, both models have the
potential to be utilized in studying the anaerobic biodegradation
kinetics of this biomass mixture. Hence, subsequent discussions
were limited to the biogas yield rate models of Moser and Hills.
It is interesting to note that the Moser and Hills growth rate
models which formed the basis for these selected models could
be described as homologues in which, the characteristic coefcient m (which is always greater than unity) differentiates them

from the Monod growth models where m is equal to unity. Moser

considered this coefcient m to be related more to adaptation of
microbial population to environmental condition through process
of mutation [40] while Liu [22] proposed that the coefcient may
well be related to cooperativity among microbial speciessubstrate
pairs. However, because bacteria grown under substrate limiting
conditions may tend to adapt through process of mutation by modication at the phenotypic and genetic levels that may lead to
improve transport for growth limiting substrate and/or improve
cooperativity among adapted microbial species [23] the views held
by these researchers may not be farfetched.
In essence, by choosing the Mosers biogas yield rate model, the
Monod half saturation constants for the acidied substrate (ks ) and
the Monod half saturation constant in volatile solid equivalent (Ks )
were estimated to be 0.1558 and 1.637 g/L respectively. This estimated Monod half saturation constant for the acidied substrate
(ks ) compares reasonably with values of 0.1430.207 g/L reported
by Barthakur et al. [11] for half saturation constant for acetate
by the methanogenic bacteria population. Also, the estimated
ks , lies within the range of 0.10.41 g/L reported by Pavlostathis
and Giraldo-Gomez [41] as half saturation constant displayed
by acetoclastic methanogens. Furthermore, the biodegradability
parameters estimated using this model revealed that the recalcitrant fraction in this biomass mixture was 0.267 of the initial
volatile solids fed, and the biodegradable fraction (1 Rf ) was 0.733
of the initial volatile solids fed. The biodegradability (1 b) was
0.1925 while the biodegradable fraction remaining (b Rf ) was
0.540 of the initial volatile solids fed at ambient temperature conditions. It is interesting to note that the sum of the biodegradable
fraction remaining at the end of experiment (b Rf ) and biodegradability potential (1 b) must be equal to the biodegradable fraction
of the feedstock volatile solids (1 Rf ).
However, by choosing the Hills based biogas yield rate model,
the Hills half saturation constant for the acidied substrate (kn )
was estimated to be 0.2288. Although, no study exist in literature
that has applied the Hills growth model in studying the kinetics
of bacteria growth after it was proposed by Liu [22], the possibility
of this type of kinetics cannot be overruled because the estimated
Hills half saturation constant (kn ) seems to bear some semblance
with the Monod half saturation constant (ks ). In addition, the Hills
half saturation constant in volatile solid equivalents of 5.34 g VS/L
obtained in this study compares reasonably to the value of 5 g VS/L
reported by Angelidaki et al. [46] for household solid waste using
the Monod growth model for the acetoclastic methanogens.
Moreover, the recalcitrant fraction (Rf ) was estimated to be
0.371 which is close to 0.400 reported by Barthakur et al. [11] and
Hashimoto [47] for cow manure alone. In addition, the biodegradable fraction (1 Rf ) was calculated to be 0.628 of the initial volatile
solids fed while the biodegradability (1 b) was calculated to be
0.3136, and the biodegradable fraction remaining in the efuent
(b Rf ) was calculated to be 0.3154 of the initial volatile solids
concentration.
Furthermore, both models seem to indicate certain degree of
bacterial adaptation for substrate degradation and cooperativity

Table 2
Parameter estimate for developed biogas yield rate models.
ks (g/L)

kn (g/L)

ki (g/L)

KS =

Biogas yield rate

models

Rmax (mL/g VS/day)

Monod based
Mosers based
Hills based
Haldane (Andrew)
based
Non-competitive
Haldane based

2.358
2.200
2.200
2.732

0.127
0.155

0.268

0.228

20.230

3.260
1.637

5.123

2.769

0.209

15.317

5.200

kS

Af (bRf )n

Kn =
(g/L)

Rf

Af(s)

RMSE

5.34

2.738
2.738

1.484
1.732
1.360
1.158

0.266
0.267
0.371
0.326

0.649
0.807
0.6864
0.787

0.161
0.2762
0.205
0.128

4.28E02
5.87E03
5.87E03
2.56E02

1.718

0.282

0.775

0.1351

2.56E02

kn
Af (bRf )n

(g/L)

Author's personal copy

O.L.Y. Momoh et al. / Biochemical Engineering Journal 79 (2013) 8493

91

Table 3
Model selection technique showing AICc and percent error for selected models.
Biogas yield rate model

Equations

Monods based

R=

Mosers based

R=

Hills based

R=

Rmax S0

(kS /Af (bRf )n )+S0

Rmax S m
0
mn

(kS /Am (bRf )

f

Rmax S m

(ksm /Am (bRf )

f

Haldane (Andrews) based

R=

Non-competitive (Haldane)

R=

)+S m

0
mn

AICC

Number of
parameters

Observed (graphical)
Ks or Kn (g/L)

Estimated Ks or
Kn (g/L)

3.500

3.200

6.000

1.637

97.33

72.71

5.500

5.340

97.33

2.91

4.000

5.123

4.000

5.200

Percent
error (%)

)+S m
0
Rmax S0

S0 +(kS /Af(s) (bRf )n )+((S 2 )(Af(s) (bRf )n )/ki )

0

Rmax S0 Af(s) (bRf )n

(kS +S0 Af(s) (bRf )n )(1+(S0 Af(s) (bRf )n /ki ))

among the acidogenic and acetogenic/methanogenic bacterial

population because n and m were greater than unity. The
coefcients of adaptation for degradation by acidogenic bacteria
(n) considering the Moser and Hills based biogas yield rate models were 1.732 and 1.360 respectively, while the coefcient of
adaptation for cooperativity by the acetogenic/methanogenic bacterial (m) was estimated to be 2.738 each for both the Moser and
Hills based biogas yield rate models. Because these coefcients
were higher than unity, some degree of bacterial adaptation and/or
cooperativity was implied. Adaptation is a necessary biological process associated with micro-organisms when grown under substrate
limiting conditions [23].
In this study, the importance of the terms n and m
cannot be overemphasized. The term m may be dened
as coefcient of adaptation for cooperativity by the acetogenic/methanogenic bacteria, while, the coefcient n can be
described as the degree of adaptation for complex biomass
degradation by the hydrolytic/acidogenic bacteria. Thus, consideration of bacteria adaptation for cooperativity m amongst the
acetogenic/methanogenic species and bacteria adaptation for complex substrate degradation n amongst the hydrolytic/acidogenic
bacteria species may contribute signicantly in the entire process
of modeling biogas yield production rate from complex biomass.
In addition to the AICc for model selection, further improvement
in model selection was conducted by comparing the percent error
between the graphically observed half saturation constant and the
half saturation constants in volatile solids equivalents estimated
through the modeling approach. The half saturation constant in
volatile solids equivalent is described as the substrate volatile solids
concentration corresponding to 0.5Rmax .The corresponding saturation constants in volatile solids equivalent are shown in Table 3
while, the combined curve tting for the tested models is shown in
Fig. 2.
Upon comparison of the percent errors, it was observed that the
Hills based biogas yield rate model provided a lower percent error
of 2.91% when compared to the 72.71% obtained from the Mosers

based biogas yield rate model (Table 3). Thus, the Hills based biogas
yield rate model may provide a reasonable description of the half
saturation constant in volatile solids equivalent (Kn ) better than the
Mosers based biogas yield rate model.
Moreover, the utilization of the linear plot similar to the socalled LineweaverBurks encountered in enzymology revealed
that, the plot of the inverse of the specic biogas yield rate obtained
at the end of the experiment (1/Re ) against the inverse of the initial
substrate volatile solids concentration (1/S0 ) yielded a linear curve
tting (Fig. 3) with slope equal to (Kn /Rmax ) and intercept equal
to (1/Rmax ). The solutions for the maximum specic biogas yield
rate (Rmax ) and half saturation constant (Kn ) were 2.3 mL/g VS/day
and 5.2 g VS/L respectively, which compare reasonably to that estimated by the Hills based biogas yield rate model than for the
Mosers biogas yield rate model.
In essence, the Hills based biogas yield rate model may be
viewed as most appropriate in studying biogas production from this
biomass mixture. By utilizing this model, the maximum specic
biogas yield rate estimated as 2.2 mL/g VS/day seem to compare
reasonably with the value of 1.75 mL/g VS/day obtained by Budiyono et al. [48] from the digestion of cow manure alone at ambient
temperature while, the substrate concentration corresponding to
this maximum biogas yield was observed at 70 g VS/L. The high rate
of biogas production from this mixture may be attributed to the
benets associated with co-digestion which include the provision
of effective buffering system, nutrient balance, microbial synergism
and reduction in toxicity linked with anaerobic digestion [8]. In
addition, the ability for the bacteria to adapt and cooperate in utilization of substrate may have strongly inuenced the rate of biogas
production as highlighted in this study.
In general, it is important to note that the Contois and two
phase kinetic models which can also be used to model hydrolysis [24] were inapplicable in this study because these models
are directly dependent on bacteria biomass concentration which
was not feasible to evaluate in the study due to the difculty

2
Monod's biogas yield rate model
1.5

Moser's biogas yeild rate model

Hill's biogas yield rate model

Haldane's (Non -Compeve)biogas yield rate model

Haldane's (Andrews)biogas yield rate model

0.5

Experimental
0
0

20

40

60

80

100

(1/specic biogas yield rate) (ml/g

VS/day)-1

specic biogas yield rate (ml/g VS/ day)

Linear ( Experimental)
2.5

0.7
0.6
0.5

0.4
0.3

y = 2.2791x + 0.4287
R = 0.9382

0.2
0.1
0
0

0.02

0.04

0.06

0.08

0.1

120

Substrate concentraon (g VS/L)

Fig. 2. Combined graphs of specic biogas yield rate against volatile solids concentration.

1/substrate concentraon (g VS/L) -1

Fig. 3. Plot of the inverse of the specic biogas yield rate (1/Re ) against the inverse
of the substrate volatile solids concentration (1/S0 ).

Author's personal copy

O.L.Y. Momoh et al. / Biochemical Engineering Journal 79 (2013) 8493

involved in differentiating between bacteria biomass and complex

biomass volatile solids. The utilization of an nth-order model of
Grau [22,23,29] as applied in this study enabled for the integration of bacteria behavior into the nth power. Also, assimilation of
acidied substrate and growth of acetogenic/methanogenic bacteria was observed to be most appropriately described by the Hills
growth model as against the Monods growth model that was original developed for pure culture utilizing homogenous substrate
[23]. Thus, there may be need to consider the Hills growth model
during anaerobic degradation of complex biomass especially for
co-digested complex substrates especially where improved biogas
yield has been reported [8].
4.2. Application of kinetic models
4.2.1. Appropriate replacement for rst order models
The approach to modeling anaerobic digestion has been grouped
into three broad categories by Tomei et al. [17]. This includes simple
substrate characterization models; intermediate substrate characterization models and advance substrate characterization models.
The simple substrate characterization models do not distinguish
between different components of the substrate into protein carbonhydrates, lipid, etc., and they are the rate limiting type models.
However, the advance substrate characterization models require
the substrate be characterized into carbonhydrates, proteins, lipids,
etc. before they can be utilized. In addition, the input parameters
needed to implement these type of models are usually numerous.
Example of advanced type models include the models developed
by Angelidaki et al. [16] and Anaerobic Digestion model no. 1 [2].
In this study, it is evident that a simple substrate characterization model has been developed that describes biogas production
from complex biomass. This modeling approach has the advantage
of providing sufcient information about the anaerobic digestion
kinetics and the nature of substrate undergoing anaerobic decomposition from very little input data. In addition, this approach
eliminates the need to quantify the viable bacteria biomass volatile
suspended solids which is usually very difcult to estimate for complex biomass [9] and a necessary requirement when utilizing the
Contois and the two phase base models [22].
Traditionally, the rst order models which are examples of simple substrate characterization models have largely been employed
in the well known biochemical methane potential assay (BMP)
and also in the design of anaerobic systems to evaluate anaerobic
biodegradability and plant design. Though, rst order models are
easy to handle, they fail to provide any information about substrate
concentration required for maximum biogas production. However,
with the modeling approach developed in this study, the biochemical methane potential assay can be evaluated in a more holistic
manner. In addition, the substrate concentration corresponding to
maximum biogas yield can easily be estimated thus, contributing
to the design and optimization of anaerobic process.
4.2.2. Determination of carbon ux and the rate limiting
coefcient (Af(s) )
In a multi-step process, the step which limits or controls the rate
of the overall process is called the rate limiting step [41]. In anaerobic digestion with its multi-step processes, the hydrolysis is usually
assumed to be the rate limiting step in the anaerobic digestion of
particulate or complex biomass [24], and the methanogenesis step
is considered to be the rate limiting step for the anaerobic digestion
of soluble substrates [40].
In this study, it was possible to utilize numeric values to
approximate the rate limiting step such that the identication of
hydrolysis/acidogenesis step was less tedious as value of (Af ) less
than 0.5 conrmed hydrolysis/acidogenesis as rate limiting step in
the anaerobic digestion of the biomass mixture.

Fractional Proportion of the maximimum

biogas yield rate (R/Rmax)

92

1.2

Af=0.1

Af=0.2

Af=0.3

Af=0.4

Af=0.5

Af=0.6

Af=0.8

Af=1

1
0.8
0.6
0.4
0.2
0

20

40
60
80
100
Volatile solids concentration (g/L)

120

Fig. 4. Fractional proportion of the maximum biogas yield rate against the volatile
solids concentration for the Hills based biogas yield rate model.

Fig. 4 shows the effect of carbon ux on biogas yield rate, in

which the fractional proportion of the maximum biogas yield rate
(R/Rmax ) was plotted against the initial volatile solids concentration
utilized in this study (S0 ).
It was observed that the volatile solids concentration within
3070 g VS/L may be appropriate for maximizing biogas production while concentration below this range was observed to reduce
biogas production from this biomass mixture at ambient temperature. Furthermore, it can be observe that conditions which tend to
restrict hydrolysis that is, reduce the value of the rate limiting coefcient or solubilization fractional efciency (Af ) (such as, during
the digestion of recalcitrant substrate) can reduce fractional biogas
production rates at low volatile solid concentration below 30 g VS/L.
On the other hand, conditions that tend to improve hydrolysis that
is, increase the value of the rate limiting coefcient (Af ), (such as,
during the digestion easy hydrolysable substrates or physiochemical pre-treatment of complex biomass) can lead to an increased
fractional biogas production rates at low volatile solids concentration below 30 g VS/L. These ndings clearly explain why mechanical
treatment of complex biomass may tend to enhance biogas production [49,50].
In general, the restriction or ease of carbon ow through the
interlinked biochemical reactions may play a crucial role in the
determination of the rate limiting step and also, biogas production rate during the anaerobic digestion of complex biomass. These
ndings tend to give credence to the works of Pavlosthatis et al. [41]
who reported that the rate limiting step may change depending on
the nature of substrate and other factors.
5. Conclusions
In this study, the co-digestion of cow manure and waste paper
(75:25) was observed to result in an increase in biogas production when compared to the digestion of these substrates alone.
The process of studying the kinetics and biodegradability of this
optimal mixture revealed that the Hills based biogas yield rate
model was most appropriate in describing the biogas production
from this mixture of complex biomass. The developed Hills based
biogas yield rate model was able to account for adaptation by
the acidogenic bacteria to degrade complex biomass (n) and also
the adaptation for cooperativity by the acetogenic/methanogenic
species (m) to assimilate acidied substrate. The half saturation
constants obtained using this model showed comparable values to
that obtained when acetate was considered growth limiting. The
biodegradability, biodegradable fraction and recalcitrant fraction
were estimated to 0.3136, 0.628 and 0.371 respectively while, the

Author's personal copy

O.L.Y. Momoh et al. / Biochemical Engineering Journal 79 (2013) 8493

rate limiting coefcient was estimated to be 0.205 implying that,

hydrolysis was rate the limiting step.
In general, this modeling approach seems to breach the gap
between the simplied rst order models and the advance substrate characterization models, and it may provide more benets
in designing of anaerobic systems as compared to the rst order
modeling approach. Additionally, the application of the modeling
approach in the biochemical methane potential assay may provide
advanced information about the biodegradability of biomass utilized in anaerobic digestion.

[22]
[23]

[24]

[25]

[26]

Acknowledgment
[27]

This research was support by the Petroleum Technology

Development Fund Local Scholarship Scheme (grant number
PTDF/TR/LS/MOLY/215/54) Nigeria.

[28]
[29]

References
[1] J.D. Batstone, Mathematical modelling of anaerobic reactor treating domestic
wastewater: rational criteria for model use, Rev. Environ. Sci. Bio/Technol. 5
(2006) 5771.
[2] J.D. Batstone, J. Keller, I. Angelidaki, S.V. Kalyuzhnyi, S.G. Pavlostathis, A. Rozzi,
W.T.M. Sanders, H. Siegrist, V.A. Vavilin, The IWA anaerobic digestion model
no 1 (ADM1), Water Sci. Technol. 45 (10) (2002) 6573.
[3] O.L.Y. Momoh, I.L. Nwaogazie, The effect of waste paper on the kinetics of biogas yield from the co-digestion of cow manure and water hyacinth, Biomass
Bioenergy 35 (2011) 13451351.
[4] H.B. Moller, S.G. Sommer, B.K. Ahring, Methane productivity of manure, straw
and solid fractions of manure, Biomass Bioenergy 26 (2004) 485495.
[5] D.P.M. Zaks, N. Winchester, C.J. Kucharik, C.C. Bardford, S. Paltser, J.M. Reily,
Contribution of anaerobic digesters of emissions mitigation and electricity
generation under U.S. Climate Policy, Environ. Sci. Technol. 45 (16) (2010)
67356742.
[6] R. Steffen, O. Szolar, R. Braun, Feed Stock for Anaerobic Digestion, Ad Nett Report
Anaerobic Digestion, Making Energy and Solving Modern Waste Problem, 2000.
[7] J.A. Alvarez, L. Otero, J.M. Lema, A method of optimizing feed composition for
anaerobic co-digestion of agro-industrial waste, Bioresour. Technol. 101 (2010)
11531158.
[8] A. Cesaro, V. Naddeo, V. Amodio, V. Belgiorno, Enhanced biogas production from
anaerobic co-digestion of solid waste by sonolysis, Ultrason. Sonochem. 19 (3)
(2012) 596600.
[9] P. Shanmugam, N.J. Horan, Simple and rapid methods to evaluate methane
potential and biomass yield for a range of mixed solid wastes, J. Bioresour.
Technol. 100 (2008) 471474.
[10] P. Mahnert, B. Linke, Kinetic study of biogas production from energy crops and
animals waste slurry: effect of organic loading rate and reactor size, Environ.
Technol. 30 (1) (2009) 9399.
[11] A. Barthakur, M. Bora, H.D. Singh, Kinetic model for substrate utilization and
methane production, in the anaerobic digestion of organic feeds, Biotechnol.
Prog. 7 (1991) 369376.
[12] D.T. Hill, Simplied Monod kinetics of methane fermentation of animal wastes,
Agric. Waste 5 (1983) 116.
[13] I. Angelidaki, L. Ellegaard, B.K. Ahring, A mathematical model for dynamic simulation of anaerobic digestion of complex substrate: focusing on ammonia
inhibition, Biotechnol. Bioeng. 42 (1993) 159166.
[14] H. Siegrist, D. Rengli, W. Gujer, Mathematical modeling of anaerobic mesophilic
sewage sludge treatment, Water Sci. Technol. 27 (1993) 2536.
[15] V.A. Vavilin, S.V. Rytov, L.Y. Lokshina, S.G. Povalostathis, M.A. Barlaz, Distributed
models of solid waste anaerobic digestion: effects of leachate recirculation and
pH adjustment, Biotechnol. Bioeng. 81 (1) (2002) 6673.
[16] I. Angelidaki, L. Ellegard, B.K. Ahring, A comprehensive model of anaerobic
bioconversion of complex substrates to biogas, Biotechnol. Bioeng. 63 (1999)
363372.
[17] M.C. Tomei, M.C. Braguglia, G. Cento, G. Mininni, Modeling of anaerobic digestion of sludge, Crit. Rev. Environ. Sci. Technol. 39 (2009) 10031051.
[18] W.T.M. Sanders, A.H.M. Veeken, G. Zeeman, J.B. Van Lier, Analysis and optimization of the anaerobic digestion of the organic fraction of municipal solid
waste, in: J. Mata Alvarez (Ed.), Bio-methanization of the Organic Fraction of
Municipal Solid Wastes, IWA Publishing, London, 2003.
[19] B. Linke, Kinetic study of thermophilic anaerobic digestion of solid wastes from
potato processing, Biomass Bioenergy 30 (2006) 892896.
[20] L.A. Fdez-Gelfo, C. lvarez-Gallego, D. Sales, L.I. Romero Garca, Drythermophilic anaerobic digestion of organic fraction of municipal solid waste:
methane production modeling, Waste Manage. 32 (2012) 382388.
[21] L.A. Fdez-Gelfo, C. lvarez-Gallego, D. Sales Mrquez, L.I. Romero Garca,
The effect of different pretreatments on biomethanation kinetics of industrial

[30]
[31]

[32]
[33]
[34]

[35]

[36]
[37]

[38]

[39]

[40]

[41]
[42]

[43]
[44]

[45]
[46]

[47]

[48]

[49]

[50]

93

Organic Fraction of Municipal Solid Wastes (OFMSW), Chem. Eng. J. 171 (2011)
411417.
Y. Liu, A simple thermodynamic approach for derivation of a general Monod
equation for microbial growth, Biochem. Eng. J. 31 (2006) 102105.
K. Kavarova-Kovar, T. Egli, Growth kinetics of suspended microbial cells from
single-substrate-controlled growth of mixed-substrate kinetics, Microbiol.
Mol. Biol. Rev. 62 (3) (1998) 646666.
J.A. Eastman, J.F. Ferguson, Solubilization of particulate organic carbon during
the acid phase of anaerobic digestion, J. Water Pollut. Control Fed. 53 (1981)
352366.
V.A. Vavilin, B. Fernandez, J. Palatsi, X. Flotats, Hydrolysis kinetics in anaerobic
degradation of particulate organic material: an overview, Waste Manage. 28
(2008) 939951.
G. Esposito, L. Frunzo, A. Panico, G. dAntonio, Mathematical modeling of
disintegration-limited co-digestion of OFMSW and sewage sludge, Water Sci.
Technol. 58 (7) (2008) 15131519.
G. Esposito, L. Frunzo, A. Panico, F. Pirozzi, Modelling the effect of the OLR and
OFMSW particle size on the performances of an anaerobic co-digestion reactor,
Process Biochem. 46 (2011) 557565.
W.T.M. Sanders, M. Geerink, G. Zeeman, G. Lettinga, Anaerobic hydrolysis kinetics of particulate substrates, Water Sci. Technol. 41 (2000) 1724.
J.K. Kim, B.R. Oh, Y.N. Chun, S.W. Kim, Effect of temperature and hydraulic retention time on anaerobic digestion of food waste, J. Biosci. Bioeng. 102 (4) (2006)
328332.
M. Faisal, H. Unno, Kinetic analysis of palm oil mill wastewater treatment by a
modied anaerobic bafed reactor, Biochem. Eng. 9 (2001) 2531.
A.A.L. Zinatizadeh, A.R. Mahamed, G.D. Najafpour, M. Isa-Hasnain, H. Nasrollahzadeh, Kinetic evaluation of palm oil mill efuent digestion in a high rate
up ow anaerobic sludge xed lm bioreactor, Process Biochem. 41 (2006)
10381046.
Y. Chen, J.J. Cheng, K.S. Creamer, Inhibition of anaerobic digestion process: a
review, Bioresour. Technol. 99 (2008) 40444064.
L.R. Droste, Theory and Practice of Waste Water Treatment, John Wiley and
Sons Inc., USA, 1997.
B. Barrios-Marrtinez, A. Marrot, P. Moulin, N. Roche, Biodegradation of high
phenol concentration by activated sludge in an immersed membrane bioreactor, Biochem. Eng. J. 30 (2006) 174183.
N. Noykova, T.G. Muller, M. Gyllenberg, J. Timmer, Quantitative analyses of
anaerobic wastewater treatment processes: identiability and parameter estimation, Biotechnol. Bioeng. 78 (1) (2001) 89103.
D. Deublein, A. Steinhauser, Biogas From Waste and Renewable Sources, 2nd
ed., Wiley-VCH Verlag GmbH & Co, Weinhein, 2011.
V.A. Vavilin, S.V. Rytov, L.Y. Lokshina, S.G. Pavlostathis, M.A. Barlaz, Distributed
model of solid waste anaerobic digestion, effects of leachate recirculation and
pH adjustment, Biotechnol. Bioeng. 81 (1) (2003) 6673.
V.A. Vavilin, I. Angelidaki, Anaerobic degradation of solid material: importance
of initiation centers for methanogenesis, mixing intensity, and 2D distributed
model, Biotechnol. Bioeng. 89 (1) (2005) 113122.
A. Dotsch, J. Severin, W. Alt, E.A. Galinski, J.U. Kreft, A mathematical model for
growth and osmoregulation in halophilic bacteria, Microbiology 154 (2008)
29562969.
M. Gerber, R. Span, An analysis of available mathematical models for anaerobic
digestion of organic substances for production of biogas, in: International Gas
union Research Conference, Paris, 2008.
S.G. Pavlostathis, E. Giraldo-Gomez, Kinetics of anaerobic treatment: a critical
review, Crit. Rev. Environ. Control 21 (56) (1991) 411490.
K.P. Burnham, D.R. Anderson, Model Selection and Multimodel Inference: A
Practical Information-Theoretic Approach, 2nd ed., Springer-Verlag, New York,
2002.
APHA, AWWA, WPCE, Standard Methods for the Examination of Water and
Wastewater, 16th ed., APHA, Washington, DC, 1985.
J. Shin, S. Han, K. Eom, S. Jung, S. Park, H. Kim, Predicting methane production
potential of anaerobic co-digestion of swine manure and food waste, Environ.
Eng. Res. 13 (2) (2008) 9397.
G. Tchobanoglous, H. Theisen, S. Vigil, Integrated Solid Waste Management
Engineering Principle and Management Issues, Mcgraw-Hill, U.S, 1993.
I. Angelidaki, X. Chen, J. Cui, P. Kaparaju, L. Ellegaard, Thermophylic anaerobic
digestion of source-sorted organic fraction of household municipal solid waste;
start up procedure for continuously stirred tank reactor, Water Res. 40 (14)
(2006) 26212628.
A.G. Hashimoto, Methane from cattle waste: effects of temperature, hydraulic
retention time, and inuent substrate concentration on kinetic parameter (k),
Biotechnol. Bioeng. 24 (9) (1982) 20392052.
I.N. Budiyono, S. Widiasa, Johari, Sunarso, The kinetics of biogas production
rate from cattle manure in batch mode, Int. J. Chem. Biomol. Eng. 3 (1) (2010)
3944.
L. Palmowski, J. Muller, Anaerobic degradation of organic materialssignicance of the substrate surface area, Water Sci. Technol. 47 (12) (2003)
231238.
J.P. Delgenes, V. Penaud, R. Moletta, Pre-treatment for the enhancement of
anaerobic digestion of solid wastes, in: J. Mata-Alvarez (Ed.), Biomethanization
of the Organic Fraction of Municipal Solid Waste, IWA Publishing, 2003.